Exercising self-control is a process that requires individuals to override or inhibit their thoughts, emotions, urges, and behaviors. Self-control failures are frequently linked with and offered as explanations for a variety of negative outcomes, including substance use. In experimental studies of self-control capacity, state levels of self-control are inferred through performance on behavioral measures thought to reflect this resource. However, constraining definitions of self-control capacity to strictly behavioral measurements limits the applicability of models of self-control to the phenomena of self-control failure as it occurs in everyday life. The proposed work makes use of self-report measures of state self-control in the context of diary and experience sampling studies to glean a more accurate representation of the variation people experience in their capacity to exhibit self-control in everyday life. Specifically, the proposed project aims to (1) understand how natural fluctuations in self-control occur by modeling trajectories of self-control capacity over time, (2) identify how the trajectory of self-control changes in response to demands on self-control, and (3) examine the trajectory of self- control characteristic of new college students-a sub-population at increased risk for substance use. Three prospective studies have been designed to address each of the aims. Study 1 will address how natural fluctuations in self-control occur (aim 1) and examine the trajectory of self-control in relation to demands on self-control (aim 2). In study 1, undergraduate students will be recruited for a diary study, in which they will be asked to report on self-control demands and state self-control several times each day for a 2-week period. Study 2 will examine how natural fluctuations in self-control occur (aim 1) and examine these fluctuations with respect to substance use behaviors (aim 3). In study 2, graduate students attending an event associated with excessive alcohol use will be recruited for an experience sampling study and asked to report on their state self- control and substance use behaviors at 20-30 randomly signaled events over a 36-hour period. In study 3, incoming first-year and third-year undergraduate students will be recruited to participate in a longitudinal diary study, in which they will be asked to report on self-control demands, state self-control, and substance use behaviors several times weekly over 4 semesters. By analyzing the data from each study using multilevel modeling, the results should provide better insight into how people's capacity for self-control fluctuates in response to various situational factors and why people often fail to exercise self-control, even when doing so is crucial to their health.